How to use the ops.conv1d function in ops
To help you get started, we’ve selected a few ops examples, based on popular ways it is used in public projects.
Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
indiejoseph / chinese-char-rnn / models / bytenet.py View on Github 
def decode_layer(self, input_, dilation, layer_no):
relu1 = tf.nn.relu(input_, name="dec_relu1_layer{}".format(layer_no))
conv1 = ops.conv1d(relu1, self.residual_channels, name="dec_conv1d_1_layer{}".format(layer_no))
relu2 = tf.nn.relu(conv1, name="enc_relu2_layer{}".format(layer_no))
dilated_conv = ops.dilated_conv1d(relu2, self.residual_channels,
dilation, self.decoder_filter_width,
causal = True,
name = "dec_dilated_conv_laye{}".format(layer_no)
)
relu3 = tf.nn.relu(dilated_conv, name="dec_relu3_layer{}".format(layer_no))
conv2 = ops.conv1d(relu3, 2 * self.residual_channels, name="dec_conv1d_2_layer{}".format(layer_no))
return input_ + conv2def _byetenet_residual_block(input_, dilation, layer_no, options, source_mask, train = True):
# input_ = layer_norm(input_, trainable = train)
relu1 = tf.nn.relu(input_, name = 'enc_relu1_layer{}'.format(layer_no))
conv1 = ops.conv1d(relu1, options['residual_channels'], name = 'enc_conv1d_1_layer{}'.format(layer_no))
# conv1 = layer_norm(conv1, trainable = train)
conv1 = conv1 * source_mask
relu2 = tf.nn.relu(conv1, name = 'enc_relu2_layer{}'.format(layer_no))
dilated_conv = ops.conv1d(relu2, options['residual_channels'],
dilation, options['encoder_filter_width'],
causal = True,
name = "enc_dilated_conv_layer{}".format(layer_no)
)
# dilated_conv = layer_norm(dilated_conv, trainable = train)
dilated_conv = dilated_conv * source_mask
relu3 = tf.nn.relu(dilated_conv, name = 'enc_relu3_layer{}'.format(layer_no))
conv2 = ops.conv1d(relu3, 2 * options['residual_channels'], name = 'enc_conv1d_2_layer{}'.format(layer_no))
conv2 = conv2 * source_mask
return input_ + conv2def encode_layer(self, input_, dilation, layer_no, last_layer = False):
options = self.options
relu1 = tf.nn.relu(input_, name = 'enc_relu1_layer{}'.format(layer_no))
conv1 = ops.conv1d(relu1, options['residual_channels'], name = 'enc_conv1d_1_layer{}'.format(layer_no))
conv1 = tf.mul(conv1, self.source_masked_d)
relu2 = tf.nn.relu(conv1, name = 'enc_relu2_layer{}'.format(layer_no))
dilated_conv = ops.dilated_conv1d(relu2, options['residual_channels'],
dilation, options['encoder_filter_width'],
causal = False,
name = "enc_dilated_conv_layer{}".format(layer_no)
)
dilated_conv = tf.mul(dilated_conv, self.source_masked_d)
relu3 = tf.nn.relu(dilated_conv, name = 'enc_relu3_layer{}'.format(layer_no))
conv2 = ops.conv1d(relu3, 2 * options['residual_channels'], name = 'enc_conv1d_2_layer{}'.format(layer_no))
return input_ + conv2def encoder(self, input_):
curr_input = input_
for layer_no, dilation in enumerate(self.self.encoder_dilations):
layer_output = self.encode_layer(curr_input, dilation, layer_no)
# ENCODE ONLY TILL THE INPUT LENGTH, conditioning should be 0 beyond that
layer_output = tf.mul(layer_output, self.source_masked, name="layer_{}_output".format(layer_no))
curr_input = layer_output
# TO BE CONCATENATED WITH TARGET EMBEDDING
processed_output = tf.nn.relu( ops.conv1d(tf.nn.relu(layer_output),
self.residual_channels,
name="encoder_post_processing") )
processed_output = tf.mul(processed_output, self.source_masked_d, name="encoder_processed")
return processed_outputdef encoder(self, input_, train = True):
options = self.options
curr_input = input_
for layer_no, dilation in enumerate(self.options['encoder_dilations']):
layer_output = self.encode_layer(curr_input, dilation, layer_no, train)
curr_input = layer_output
processed_output = tf.nn.relu( ops.conv1d(tf.nn.relu(layer_output),
options['residual_channels'],
name = 'encoder_post_processing') )
return processed_outputdef decoder(self, input_, encoder_embedding = None):
options = self.options
curr_input = input_
if encoder_embedding != None:
# CONDITION WITH ENCODER EMBEDDING FOR THE TRANSLATION MODEL
curr_input = tf.concat(2, [input_, encoder_embedding])
print "Decoder Input", curr_input
for layer_no, dilation in enumerate(options['decoder_dilations']):
layer_output = self.decode_layer(curr_input, dilation, layer_no)
curr_input = layer_output
processed_output = ops.conv1d(tf.nn.relu(layer_output),
options['n_target_quant'],
name = 'decoder_post_processing')
return processed_outputdef decode_layer(self, input_, dilation, layer_no):
options = self.options
relu1 = tf.nn.relu(input_, name = 'dec_relu1_layer{}'.format(layer_no))
conv1 = ops.conv1d(relu1, options['residual_channels'], name = 'dec_conv1d_1_layer{}'.format(layer_no))
relu2 = tf.nn.relu(conv1, name = 'enc_relu2_layer{}'.format(layer_no))
dilated_conv = ops.dilated_conv1d(relu2, options['residual_channels'],
dilation, options['decoder_filter_width'],
causal = True,
name = "dec_dilated_conv_laye{}".format(layer_no)
)
relu3 = tf.nn.relu(dilated_conv, name = 'dec_relu3_layer{}'.format(layer_no))
conv2 = ops.conv1d(relu3, 2 * options['residual_channels'], name = 'dec_conv1d_2_layer{}'.format(layer_no))
return input_ + conv2def decode_layer(self, input_, dilation, layer_no):
options = self.options
relu1 = tf.nn.relu(input_, name = 'dec_relu1_layer{}'.format(layer_no))
conv1 = ops.conv1d(relu1, options['residual_channels'], name = 'dec_conv1d_1_layer{}'.format(layer_no))
relu2 = tf.nn.relu(conv1, name = 'enc_relu2_layer{}'.format(layer_no))
dilated_conv = ops.dilated_conv1d(relu2, options['residual_channels'],
dilation, options['decoder_filter_width'],
causal = True,
name = "dec_dilated_conv_laye{}".format(layer_no)
)
relu3 = tf.nn.relu(dilated_conv, name = 'dec_relu3_layer{}'.format(layer_no))
conv2 = ops.conv1d(relu3, 2 * options['residual_channels'], name = 'dec_conv1d_2_layer{}'.format(layer_no))
return input_ + conv2def encode_layer(self, input_, dilation, layer_no, last_layer = False):
options = self.options
relu1 = tf.nn.relu(input_, name = 'enc_relu1_layer{}'.format(layer_no))
conv1 = ops.conv1d(relu1, options['residual_channels'], name = 'enc_conv1d_1_layer{}'.format(layer_no))
conv1 = tf.mul(conv1, self.source_masked_d)
relu2 = tf.nn.relu(conv1, name = 'enc_relu2_layer{}'.format(layer_no))
dilated_conv = ops.dilated_conv1d(relu2, options['residual_channels'],
dilation, options['encoder_filter_width'],
causal = False,
name = "enc_dilated_conv_layer{}".format(layer_no)
)
dilated_conv = tf.mul(dilated_conv, self.source_masked_d)
relu3 = tf.nn.relu(dilated_conv, name = 'enc_relu3_layer{}'.format(layer_no))
conv2 = ops.conv1d(relu3, 2 * options['residual_channels'], name = 'enc_conv1d_2_layer{}'.format(layer_no))
return input_ + conv2
